Journal
Appl
Journal of Applied Horticulture, 22(1): 76-79, 2020
Journal
of Applied
Horticulture
DOI: 10.37855/jah.2020.v22i01.15
ISSN: 0972-1045
Stability analysis and genotype x environment interaction for
flower yield and quality traits in marigold (Tagetes spp.)
M.A. Patel*, S.L. Chawla and S.T. Bhatt
Department of Floriculture and Landscape Architecture, ASPEE College of Horticulture and Forestry, Navsari Agricultural
University, Navsari - 396 450, Gujarat, India. *E-mail: mukeshpatel@nau.in
Abstract
The stability analysis of 26 diverse genotypes of marigold (Tagetes spp.) carried out over three different environments, revealed that the
differences among genotypes and environments were highly significant for all the characters when tested against both pooled error as
well as pooled deviation. The analysis further revealed that component of G x E (linear) had most contribution for plant height, number
of secondary branches per plant, days to first flowering, flower diameter, flower weight, number of flowers per plant and flower yield
per plant indicating significant differences among the genotypes for their regression on environmental indices. Considering the three
stability parameters, Local Selection 13 for flower yield (414.40 g/plant); F1 White Dwarf, Local Selection 2 and Namdhari African
Orange for earliness and Local Selection 9, Local Selection 14 and Local Selection 13 for individual flower weight were identified as
promising genotypes for further improvement programme.
Key words: Marigold, stability, genotypes, environments
Introduction
Marigold (Tagetes spp.) is one of the most important traditional
flower crops grown in India, owing to its ornamental and
industrial uses. It is grown for landscaping and occupies an
ever increasing demand in medicinal and industrial sector. It is
widely grown for its loose flowers used for religious offerings and
making garlands during social functions and as a bedding plant in
landscape gardening. The extraction of carotenoides from petals
for industrial uses raised the importance of this crop and increased
the area under its cultivation. Marigold is also suggested as trap
crop for monitoring the Helicoverpa incidence in vegetable crops
and has nematicidal properties also.
The loose flowers are regarded as the backbone of the Indian
floriculture industry as it holds the major share in area and
production. In breeding programme, it is necessary to screen and
develop stable genotypes which perform more or less uniform
under varying environmental conditions. Thus, knowledge of
genotype x environment interactions helps the breeder to select
high yielding and more adaptable varieties or hybrids. Looking
at the importance and commercial potential, there is an urgent
need to identify potential genotypes, which would result in further
improvement and to develop cultivar for specific uses.
Keeping this background in view, the present study was
undertaken to assess the performance of genotypes over
environments.
Materials and methods
The investigation was carried out during Rabi season (November
2014 - April 2015) at three different locations, viz., Floriculture
Research Farm, Navsari (Dist. Navsari), Regional Rice Research
Station, Vyara (Dist. Tapi), Hill Millet Research Station, Waghai
(Dist. The Dangs) of Navsari Agricultural University (Gujarat).
Twenty six (26) genotypes collected from diverse source
were grown in a randomized block design (RBD) with three
replications. The 26 genotypes of marigold comprised four F1
hybrids (Inca Gold, Inca Yellow, F1 White Dwarf and Sonata
Orange); fifteen local genotypes (Local Sel. 1, Local Sel. 2, Local
Sel. 3, Local Sel. 4, Local Sel. 5, Local Sel. 6, Local Sel. 7, Local
Sel. 8, Local Sel. 9, Local Sel. 10, Local Sel. 11, Local Sel. 12,
Local Sel. 13, Local Sel. 14 and Local Sel. 15) as well as seven
open pollinated varieties (Pusa Narangi Gainda, Summer Sugat,
Namdhari African Orange, Hawaii Orange, Swati Orange, Indus
Orange Bunch and Suvarna Orange).
Seeds of all the genotypes were sown on the raised beds in the
month of November to raise seedlings. Transplanting of seedlings
was done when they attained three to four true leaves stage. The
genotypes were planted in a single-row of 20 plants under each
replication with a spacing of 60 x 45 cm with all the recommended
agronomical practices and plant protection measures. The
observations were recorded on five randomly tagged plants from
each genotype of each replication. For all the characters under
study, the mean values of five randomly selected plants were
calculated for each observation under individual location. The
analysis of variance representing the mean square due to different
sources of variation for various traits including magnitude of G x
E interactions and stability parameters were estimated as per the
procedure outlined by Eberhart and Russell (1966) to understand
the genotype x environment interactions of different genotypes
and to assess stability of individual genotype. The mean sums
of squares for phenotypic stability for different characters are
presented in Table 1.
Journal of Applied Horticulture (www.horticultureresearch.net)
Stability analysis and genotype x environment interaction for flower yield and quality traits in marigold
77
Table 1. Analysis of variance for phenotypic stability pertaining to various characters in marigold (pooled)
Source
d.f.
Plant
height
(cm)
Plant
spread
(cm)
Number
Days
of primary
to
branches/
first
plant
flowering
Duration
of
flowering
(days)
Flower
diameter
(mm)
Flower
weight
(g)
Number
of
flowers /
plant
Flower
yield /
plant
(g)
Genotypes (G)
25
631.64 **++
190.32**++
4.28**++
132.17**++
738.68**++
319.18**++
15.28**++
201.39**++
48596.40**++
Environments (E)
2
10487.35**++ 1358.35**+
32.51**+
262.28*
2218.75**+
2992.00**+
103.48**+
4650.70**+ 936241.33**++
Environments (Lin.) 1
20974.70**++ 2716.69**++
65.02**++
524.55**++
4437.50**++ 5984.01**++
206.95**++
9301.41**++ 1872482.66**++
GxE
50
99.86**++
21.35**
0.77**
30.21**+
67.45**
28.89**++
3.86**++
58.84**+
17532.69**++
G x E (Lin.)
25
171.93**++
23.67**
0.83**
40.31**+
78.37**
45.86**++
5.65**++
77.23**+
29336.18**++
Pooled deviation
26
26.72
18.29**
0.68**
19.33**
54.35**
11.46
1.99**
38.89**
5508.85**
Pooled error
150
18.75
2.80
0.21
5.27
9.64
12.59
0.25
2.50
716.59
*, ** Significant at 5 % and 1 % level, respectively against pooled error.
+, ++ Significant at 5 % and 1 % level, respectively against pooled deviation.
Results and discussion
Analysis of variance for phenotypic stability: The results
of stability analysis over three locations presented in Table 1
revealed that all the genotypes differed significantly among
them and exhibited significant differences when tested against
both pooled error and pooled deviation for all the characters
under study as well as significant variation existed among the
environments (except days to first flowering) when tested against
pooled deviation. Similar results have been reported by Naik et
al. (2005) in marigold, Sreenivasulu et al. (2007) in china aster
and Kirtimala et al. (2011) in gladiolus.
The genotype x environment (G x E) components were also found
significant to highly significant for most of the characters viz., plant
height, days to first flowering, flower diameter, flower weight,
number of flowers per plant and flower yield per plant when tested
against pooled error and pooled deviation, which satisfied the
requirement of stability analysis suggesting differential reaction
of genotypes to varied environments. Therefore, stability for these
characters was tabulated. Significance of G x E interaction has
also been reported by Nalawadi (1982) in marigold, Desh Raj
and Misra (1998) in gladiolus and Misra and Gupta (2005) in
carnation, Sreenivasulu et al. (2007) in china aster and Kirtimala
et al. (2011) in gladiolus.
for their regression on environmental indices. Pooled deviation
were significant for all the characters when tested against
pooled error suggesting that deviation for linear regression also
contributed substantially towards the differences in stability of
genotypes. However, considerable portion of G x E interactions
was attributable to linear components (predictable) for plant
height, days to first flowering, flower diameter, flower weight,
number of flowers per plant and flower yield per plant. Thus, this
study indicated that both linear and non-linear functions play
an important role in building up total G x E interaction. These
findings are in agreement with the results of Desh Raj and Misra
(1998) in gladiolus, Sreenivasulu et al. (2007) in china aster and
Kirtimala et al. (2011) in gladiolus.
Stability parameters: The mean, regression coefficient (bi)
and mean square deviation from linear regression line (S2di)
are the three stability parameters proposed by Eberhart and
Russel (1996) in their stability model. The parameters estimated
to evolve related stability of different population types over a
range of environmental conditions are presented character wise
in Table 2 and 3.
A variety having good adaptability is one that consistently gives
stable performance over wide range of environments (Frey, 1964).
Thus, stability depends upon the relative sensitivity of a genotype
to varied environments. An individual may react to variable
environments in such a way that its development is buffered
against environmental variation and the same adaptive phenotype
being produced in varying environments. This situation has
been named as canalization (Waddington, 1942); developmental
stability (Mather, 1943); phenotypic stability (Lewis, 1954) and
developmental homeostasis (Lerner, 1954).
Eberhart and Russell (1966) defined stable genotypes as one
which showed a high mean yield, regression coefficient (bi)
around unity and deviation from regression (S2di) equal to
zero. Later on, Breese (1969) advocated that linear regression
(bi) could be simply be regarded as measure of response of a
particular genotype, whereas the deviation from regression (S2di)
as a measure of stability. If regression co-efficient (bi) is greater
than unity with high mean values, the genotype is considered
to possess below average stability and is highly sensitive to
environments. Regression co-efficient (bi<1) with high mean
values indicates above average stability and can adapt to poor
environments. If regression co-efficient (bi) is equal to unity (bi
= 1) with high mean values, this indicates average sensitivity to
environmental changes and adaptation to various environments.
Mean squares due to environment (linear) were highly significant
for all the characters, which indicated considerable differences
among the environments and their predominant effects on
these characters. The linear response of genotype to varying
environments showed significant mean squares for all the
characters indicating significant differences among the genotypes
Among the 26 genotypes, Local Selection 15 (70.29 cm), Local
Selection 14 (70.11 cm), Local Selection 12 (64.91 cm) and
Local Selection 13 (64.38 cm) registered higher mean value for
plant height, bi value around unity and lower S2di value. These
findings suggested that these genotypes were stable and showed
predictable performance under various environments.
Journal of Applied Horticulture (www.horticultureresearch.net)
78
Stability analysis and genotype x environment interaction for flower yield and quality traits in marigold
Table 2. Stability parameters of individual genotype for different characters
Genotype
Plant height
(cm)
Number of secondary
branches per plant
Days to first flowering
(days)
Flower weight
(g)
Mean
bi
S²di
Mean
bi
S²di
Mean
bi
S²di
Mean
bi
S²di
Pusa Narangi Gainda
57.07
1.06
-12.28
21.62
0.61
-0.86
57.42
1.03
-5.08
6.75
0.60
0.01
Summer Sugat
65.58
1.25
-10.91
20.80
0.62
-0.19
55.29
1.71
-4.95
7.83
1.37
0.22
Namdhari African Orange
57.89
0.96
-18.56
26.40
0.94
74.77 **
56.42
0.53
-4.85
6.60
0.23
-0.03
Hawaii Orange
58.73
1.34
303.38 ** 17.80
0.14
11.72 **
48.38
-0.38
31.06 **
6.21
0.44
-0.07
Swati Orange
74.78
1.21
89.02 *
32.07
1.14
8.22 *
54.02
1.75
31.53 **
7.46
0.70
-0.14
Indus Orange Bunch
60.93
1.04
-18.51
31.78
1.52
0.29
57.56
-0.02
-3.90
7.24
0.92
-0.10
Local Selection 1
71.89
1.22
-16.96
35.20
1.13
-0.58
57.58
2.28
-2.84
10.81
0.83
11.82 **
Suvarna Orange
48.00
-0.18*
-18.59
35.40
1.74*
-1.44
57.58
1.92
16.10 *
6.69
0.34
0.36
F1 White Dwarf
28.07
0.22
0.88
14.82
0.61
-0.96
45.82
0.33
0.00
5.58
0.15
0.02
Local Selection 2
64.56
0.91
-13.01
31.29
1.26
1.47
55.93
1.07
-3.24
10.00
1.84
3.42 **
Local Selection 3
67.13
1.31
-17.07
36.96
1.28*
-1.37
59.31
2.64
-4.27
9.10
2.44
0.38
Local Selection 4
71.98
1.37
-12.93
32.16
1.20
22.27 **
53.67
1.87
30.85 **
10.70
2.69
6.69 **
Local Selection 5
76.58
1.74
-5.26
31.60
1.39
1.97
75.47
-2.78 240.33 ** 11.51
1.85
4.73 **
Local Selection 6
70.23
0.96
5.33
32.67
1.04
3.46
62.47
2.76
-4.60
11.08
2.43
0.14
Local Selection 7
72.42
1.44
10.46
27.40
0.62
2.28
62.62
2.00
-2.09
12.59
1.78
8.0 **
Local Selection 8
69.89
1.30
-8.47
34.42
1.40
-0.98
54.09
-0.36
3.71
11.75
0.67
4.11 **
Local Selection 9
66.82
1.20
-16.39
30.78
1.10
1.85
61.33
0.34
-4.54
12.10
0.44
-0.16
Inca Gold
30.27
0.32*
-18.51
19.82
0.82
10.15 **
43.36
-0.17
25.48 *
9.95
1.31
3.46 **
Local Selection 10
62.53
1.33
5.68
37.22
0.61
1.04
64.87
1.03
-4.92
6.29
0.47
0.11
Local Selection 11
73.84
1.40
-17.31
29.64
0.94
0.33
61.78
3.15
-4.19
8.74
1.49
0.29
Local Selection 12
64.91
1.02
40.75
32.71
0.95
-1.36
56.42
-0.67
17.60 *
7.73
0.16*
-0.22
Local Selection 13
64.38
1.04
-13.82
32.96
1.11
3.22
56.49
3.14
0.32
9.84
0.60
0.43
Inca Yellow
32.91
0.33*
-18.32
19.42
0.82
3.52
53.98
1.36
-1.06
12.27
0.07
0.32
Local Selection 14
70.11
1.01
-9.33
33.98
1.13
-0.69
60.73
-1.13
-4.10
10.58
0.41
0.04
Sonata Orange
32.51
0.13
7.08
20.16
0.69*
-1.45
45.62
1.82
5.55
8.47
2.14
1.36 *
Local Selection 15
70.29
1.07
-5.36
35.33
1.21
0.66
52.24
0.82
22.15 *
12.82
-0.34
-0.07
Population Mean
60.93
29.02
The genotype F 1 White Dwarf found superior over other
genotypes for earliness with desirable bi value almost unity and
desirable S2di values. But, Local Selection 2 and Namdhari
African Orange also showed earliness with bi values near unity
and less than unity and desirable S2di values. In case of flower
diameter, Local Selection 15 and Local Selection 2 registered
higher mean value with low S2di values with bi values near unity.
56.56
9.26
Local Selection 6 registered higher mean value for flower weight
with desirable bi value above unity along with low S2di value,
while Local Selection 9, Local Selection 14 and Local Selection
13 registered higher mean values with bi values below unity and
lower S2di values. Whereas, Pusa Narangi Gainda registered
higher mean value with bi value less than one and low S2di value.
yield per plant (414.40 g/plant) and bi value (1.16) approaching
unity together with comparatively low S2di (-589.50) value.
Therefore, Local Selection 13 may be considered superior over
the remaining genotypes under varying environments. Whereas,
Local Selection 15 exhibited high mean flower yield (575.40 g/
plant) with bi value (0.97) nearer to unity but, it had significant
value of S2di (16431.2*). This genotype although, unstable
according to the model of Eberhart and Russel (1966) however
it is of economic interest due to their specific suitability for
favourable environments. This genotype can be considered as less
adoptive and can be further evaluated for their yield performance
over more locations and years.
In the present investigation, only one genotype Local Selection 13
possessed high per se performance (over general mean) for flower
The present study revealed that the local genotypes as a group
had slightly higher mean value for flower yield per plant.
Journal of Applied Horticulture (www.horticultureresearch.net)
Stability analysis and genotype x environment interaction for flower yield and quality traits in marigold
79
Table 3. Stability parameters of individual genotype for different characters
Genotype
Number of flowers per plant
Flower yield/plant (g)
Mean
bi
S²di
Mean
bi
S²di
Pusa Narangi Gainda
39.87
0.57
5.24
222.20
0.34
2456.9 *
Summer Sugat
34.73
1.32
16.42 **
294.20
1.35
-527.80
Namdhari A.Orange
33.89
1.49
4.93
212.20
0.68
-463.20
Hawaii Orange
24.20
0.212*
-2.40
144.90
0.17*
-677.70
Swati Orange
36.58
1.39
-1.13
273.60
0.93
565.90
Indus Orange Bunch
30.51
1.38*
-2.17
229.30
0.99
-169.60
Local Selection 1
44.42
1.53
19.22 **
488.00
1.59
15713.6 **
Suvarna Orange
30.11
0.59
-1.09
196.60
0.35
144.40
F1 White Dwarf
21.02
0.86
3.47
115.50
0.30*
-629.20
Local Selection 2
44.20
1.17
189.32 **
453.00
1.59
19400.9 **
Local Selection 3
37.69
1.67
151.97 **
410.80
2.32
5123.8 **
Local Selection 4
35.60
1.24
32.04 **
427.30
2.00
7512.9 **
Local Selection 5
27.07
0.52*
-2.29
306.00
0.88
3239.4 *
Local Selection 6
50.67
1.34
234.71 **
567.70
2.31
5598.3 **
Local Selection 7
37.44
1.29
67.67 **
480.60
1.84
25774.0 **
Local Selection 8
29.87
0.80
115.73 **
359.90
0.86
17791.4 **
Local Selection 9
36.96
0.72
24.11 **
433.50
0.67
4765.5 **
Inca Gold
25.71
0.53
3.61
267.30
0.78
2332.9 *
Local Selection 10
33.00
0.76
-1.74
199.40
0.50*
-613.90
Local Selection 11
37.36
1.45
31.29 **
348.50
1.52
1846.00
Local Selection 12
39.00
1.32
12.33 *
267.30
0.61
-363.30
Local Selection 13
44.33
1.39
16.28 **
414.40
1.16
-589.50
Inca Yellow
19.22
0.36
-0.62
232.20
0.23
96.70
Local Selection 14
25.89
0.23
22.18 **
263.10
0.28
762.60
Sonata Orange
24.96
0.37
0.26
211.60
0.80
-688.80
Local Selection 15
47.27
1.52
7.91 *
575.40
0.97
16431.2 **
Population Mean
34.29
322.87
References
Breese, E.L. 1969. The measurement and significance of genotypeenvironment interactions in grasses. Heredity, 24: 27-44.
Desh Raj and R.L. Misra, 1998. Stability analysis in gladiolus-II: Floral
characters. J. Orn. Hort., 1(2): 39-44.
Eberhart, S.A. and W.A. Russel, 1966. Stability parameters for comparing
varieties. Crop Science, 6: 36-40.
Frey, K.J. 1964. Adaptation reaction of oat strains selected under stress
and non-stress environmental conditions, Crop Sci., 4: 55-58.
Kirtimala Naik, S.K. Nataraj, B.S. Kulkarni and B.B. Reddy, 2011.
Stability analysis for earliness and corm characters in gladiolus
(Gladiolus hybridus Hort.). J. Hort. Sci., 6(1): 41-45.
Naik, B.H., A.A. Patil and N. Basavaraj, 2005. Stability analysis in
African marigold (Tagetes erecta L.) genotypes for growth and
flower yield. Kar. J. Agri. Sci., 18(3): 758-763.
Lerner, I.M. 1954. Genetic Homeostasis. John Wiley & Sons, Inc.,
New York.
Lewis, D. 1954. Gene-environment interaction. A relationship between
dominance, heterosis, phenotypic stability and variability. Heredity,
8: 333-356.
Mather, K. 1943. Polygenic inheritance and natural selection. Biol.
Rev., 18: 32-64.
Misra, S. and Y.C. Gupta, 2005. Stability analysis in carnation. Prog.
Hort., 37(2): 406-411.
Nalawadi, S.C. 1982. Nutritional Studies in some Varieties of Marigold
(Tagetes erecta L.). Ph.D. Thesis Submitted to University of
Agricultural Sciences, Bangalore.
Sreenivasulu, G.B., B.S. Kulkarni, S.K. Natraj, B.S. Reddy, Kirti Mala
Naik and K. Chandan, 2007. Analysis of variance for different traits
of china aster (Callistephus chinensis L.). The Asian J. Hort., 2(2):
68-70.
Waddington, C.H. 1942. Canalization of development and the inheritance
of acquired characters. Nature, 150: 563-565.
Received: June, 2019; Revised: October, 2019; Accepted: December, 2019
Journal of Applied Horticulture (www.horticultureresearch.net)